In general, the PI will attempt to define multiple endo-/exocytotic compartments in 3T3-L1 adipocytes including vesicles which transport membrane associated proteins like GLUT4, secreted proteins where secretion is stimulated by insulin such as ARCP30 (adipocyte complement related protein of 30 kDa-cloned by the PI from a subtractive cDNA library enriched in mRNAs induced during adipocyte differentiation), and secreted proteins not stimulated by insulin like adipocyte-specific (3 VI collagen (which, unlike ubiquitously expressed type VI collagen has a 100 AA insert in the C3 domain identified by the PI using subtractive antibody screening. This technique involves immunodepleting antisera raised against secretory proteins from 3T3-L1 adipocytes of antibodies cross-reacting with proteins found in fibroblast supernatant. The PI will examine the involvement of signal transduction systems in regulating ARCP30 and GLUT4 vesicular traffic, in particular those utilized by insulin and TNFalpha. Finally, the PI will attempt to clone and characterize all adipocyte secretory proteins using his subtractive antibody technique. There are 10 specific aims. In specific aim 1, he will develop modified versions of ARCP 30 containing N-linked glycosylation sites and epitope tags to study biosynthesis, organelle processing, and insulin stimulated secretion of ARCP 30. In specific aim 2, he will use confocal microscopy and subcellular fractionation to study the intracellular location and protein composition of vesicles containing GLUT4, ARCP30, and adipocyte-specific alpha3VI collagen. In the third specific aim, the PI will determine whether global or compartmentalized changes in cytosolic calcium affect exocytosis and secretion of insulin sensitive and insensitive secretory proteins. If so, all known isoforms of synaptotagmin and phospholipase C would be immunolocalized, and exocytosis would be examined in cells expressing dominant-negative constructs. In addition the PI will study differential regulation of GLUT4, ARCP30, and alpha3VI collagen exocytosis by TNFalpha, IL-6, and Beta-adrenergic agonist. In specific aims 4 through 6, signal transduction systems will be identified which mediate insulin-stimulated exocytosis of ARCP30 and GLUT4 vesicles. The PI will introduce anti-sense constructs for Rab3A, Rab3D, caveolins 1, 2, and 3, PKC epsilon, PKC zeta, osteonectin, and adipocyte alpha3 VI collagen. Overexpression of wild type proteins will be assessed for p85 and p110 subunits of PI3-kinase, Akt, PKCepsilon, PKCzeta, PIP kinase, dynamin l and 2, caveolins 1, 2, and 3, and Rabs 3A, 3D, and 4. Dominant negative constructs for dynamin 1, Rabs 3D, 3A, and 4, and caveolin (as yet undeveloped), and constitutively active forms of Ak+, PKCepsilon, and PKCzeta will also be studied. In all these transfected cells, the PI will examine cross talk between TNFalpha and insulin receptors. These experiments will require the PI to complete development of a retroviral vector system for reproducible high level expression of recombinant proteins in 3T3-L1 adipocytes, and for inducible expression using a tetracycline-inhibitable promoter. The PI is also developing a method for assessing GLUT4 appearance on the plasma membrane which involves fluorescent cell sorting for a exofacially-tagged GLUT4. In specific aims 7 and 8, the PI states he will study the role of secreted matrix proteins, and including osteonectin, type VI collagen, integrins, laminin, and fibronectins in 3T3-L1 adipocyte differentiation and function. In addition, the PI states he will assess two novel secreted proteins containing sulfated oligosaccharides. However, only studies involving adipocyte alpha3 VI collagen are described in the experimental design section. These studies on alpha3 VI collagen include effects of insulin, TNFalpha, and IL6 on expression, pulse-chase studies of biosynthesis, immunostaining, expression of anti-sense and dominant negative constructs, effects of antibodies directed to the adipocyte specific 100-AA insert in the C3 domain, and possibly transgenic mice with deleted 100-AA insert. In specific aims 9 and 10, the PI will identify all adipocyte-specific secretory proteins using a subtractive antibody technique, and clone these proteins by direct protein microsequencing or screening expression libraries. These studies will involve 3T3-L1 cells where antisera raised against adipocyte secreted proteins will be depleted of antibodies against fibroblast secreted proteins. Also, antibodies will be raised against proteins secreted by primary cultured mouse adipocytes in the presence and absence of insulin or TNFalpha and subtracted using 3T3-L1 fibroblast and/or adipocyte secreted proteins. Functional studies and knock-out mice will be generated for each of these secreted proteins as warranted.